DEVELOPMENT OF PAVEMENT PERFORMANCE PREDICTION MODELS FOR GEORGIA PAVEMENTS. FINAL REPORT

The objective of this study was to develop a set of pavement performance prediction models for flexible pavement in the State of Georgia. The Georgia Department of Transportation (GDOT) has a flexible pavement rating system called PACES. Data collected over the past years from the PACES rating surveys has created a good pavement condition database. The GDOT has also maintained databases such as traffic, road history, and road characteristics. The data available in these databases and data from a previous research project were utilized for this study. Model development was started by selecting four basic models from the literature and modifying them to suit the basic form of PACES rating versus traffic behavior. In all four models, the dependent and independent variables were the PACES rating (PR) and the cumulative 18 kip ESAL traffic since the last rehabilitation (T18), respectively. Model development was divided into two phases. In phase I, linear regression analysis and a special technique were used to develop four sets of family curves. These four sets of family curves were compared and the best set of family curves was selected as the performance prediction model for phase I. Phase II analysis was designed to correlate pavement performance to structural, material, and environmental properties. This analysis did not find any significant correlation between these properties and pavement performance. Therefore, the prediction model developed in phase I was selected as the final prediction model. Model applications, accomplished by using a few sets of actual data, proved the effectiveness of the models. This study led to the identification of the need to upgrade the GDOT existing databases.